Blue Copper Proteins

Pia «tocyaniit-lagt Aiurxti-larn Paeudoa* unit-Spat Amicyanin-laac
PlBfftocyùnin-lpca BnaticyEnin-lrcy CHP-Ädsp stell&cyaniiv-l jer

large acidic patch

Fig. 5. Molecular electrostatic potentials for eight proteins representative of the different blue copper protein subfamilies (top) and for the wt plastocyanin and its mutants in the eastern site at 150 mM (bottom). Isopotential contours are at -0.5 (red) and +0.5 (blue) kcal/mol/e are displayed.

Q88E D42N E43N E43K E43Q/D44N

Fig. 5. Molecular electrostatic potentials for eight proteins representative of the different blue copper protein subfamilies (top) and for the wt plastocyanin and its mutants in the eastern site at 150 mM (bottom). Isopotential contours are at -0.5 (red) and +0.5 (blue) kcal/mol/e are displayed.

Fig. 6. Clusterization of the oxidized forms of the proteins listed in Table 1 in the MEP SI space. The picture is obtained by 3D projection of d(a,b) for MEPs computed at an ionic strength of 150 mM. Each point represents a protein.

otic plastocyanins and is probably important for protein binding, as indicated by the interaction of plastocyanin with cytochrome-f in the pc/cyt f complex structure (21), and (3) the region at the interface between amicyanin and its partner MADH in the X-ray ternary complex determined by Chen et al. (20). Investigation of the MIFs at these regions may reveal whether they are commonly involved in molecular recognition processes in the cupredoxin subfamilies.

2.3. Classification of the Descriptors

The different pairwise indices obtained for the MIFs were analyzed in a manner similar to the sequence identity indices. Distance matrices were computed from distances given by d(a,b) = sqrt(l-MIFSI) (7)

where MIFSI is either the molecular electrostatic potential similarity index (MEP SI) or the hydro-phobic potential similarity index (HYD SI). d(a,b) can be considered as the distance between any two proteins a and b in the molecular interaction field similarity space. Cluster analyses were then performed with these derived descriptors.

Among the different MIF similarity descriptors obtained for the protein systems, MEP SIs computed at an ionic strength of 150 mM are collected in Table 3. These values are reported here by way of an example, as they proved to be particularly informative and revealing. A complete analysis of all the collected indices is described in De Rienzo et al. (19). The results of the cluster analysis of the computed data in Table 3 are displayed in Fig. 6. This figure shows how the different proteins (in their oxidized state) cluster with respect to their molecular electrostatic properties. It is evident that the selected proteins cluster quite differently in MEP space (Fig. 6) and in sequence space (Fig. 4).

The correlations between the pairwise sequence identity percentages [ID(a,b)] and the pairwise MIF SI descriptors (both MEP SI and HYD SI) are shown in Fig. 7. From these plots, we can immediately see that HYD SIs can be approximately deduced from ID(a,b). That is to say, that hydropho-bic properties seem to be strictly connected with sequence features. This might be the result of the fact that hydrophobic effects are effective at a closer distance to the protein surface than electrostatic effects. Nevertheless, analysis of the HYD SIs clearly showed the presence of a conserved hydropho-

Table 3

Molecular Electrostatic Potential Pairwise Similarity Index [SI(a,b)] Values Computed at 150 mM on the Whole Skin of the Oxidized Form of the Proteins in Table 1

laac ladw Iag6 larn lb3i Ibaw liuz ijer Ijoi lkdj lnin Inwo lpcs Iplb Ipmy Ipnd lrcy Irkr lzia 2aza 2cbp 2plt 4azu 7pcy 8paz 9pcyaminmr laac 1 0.693 -0.057 0.077 0.636 0.542 -0.014 0.367 0.158 -0.207 0.509 0.116 0.564 -0.022 0.611-0.136 0.301 0.374 0.811 0.115 -0.114 -0.186 0.264 -0.036 0.736 -0.2 0.531

ladw 0.693 1 0.051-0.057 0.421 0.441 -0.02 0.219 0.221 0.039 0.313 0.11 0.395 0.051 0.55 -0.044 0.224 0.05 0.778 -0.059 -0.399 -0.18 -0.035 -0.048 0.853-0.113 0.685

lag6 -0.057 0.051 1 -0.062 -0.215 0.173 0.801 0.516 0.042 0.372 -.244 -0.107 0.245 0.876 -0.395 0.828 -0.087 -0.028 -0.219 0.001 -0.491 0.786 0.246 0.79 -0.212 0.877 -0.131

larn 0.077 -0.057 -0.062 1 0.398 0.343 0.186 0.184 0.418 -0.142 0.448 0.466 0.318 -0.066 0.094 -0.171 0.321 0.592 -0.035 0.564 0.339 0.137 0.446 0.092 -0.086-0.026 -0.057

lb3i 0.636 0.421 -0.215 0.398 1 0.447 -0.115 0.171 0.474 -0.385 0.616 0.353 0.45 -0.111 0.625 -0.183 0.565 0.502 0.59 0.18 0.037 -0.171 0.173 -0.075 0.499-0.173 0.252

lbaw 0.542 0.441 0.173 0.343 0.447 1 0.202 0.31 0.206 0.026 0.466 0.062 0.625 0.168 0.315 0.024 0.558 0.33 0.376 0.314 -0.332 0.004 0.364 0.053 0.374 0.001 0.35

liuz -0.014 -0.02 0.801 0.186 -0.115 0.202 1 0.642 0.25 0.316 0.017 0.152 0.11 0.799 -0.266 0.74 0.018 0.028 -0.198 0.085 -0.306 0.883 0.427 0.936 -0.291 0.734 -0.189

Ijer 0.367 0.219 0.516 0.184 0.171 0.31 0.642 1 0.142 0.262 0.047 0.159 0.445 0.556 0.163 0.495 0.244 0.235 0.149 -0.097 -0.219 0.638 0.509 0.66- 0.08 0.514 -0.056

ljoi 0.158 0.221 0.042 0.418 0.474 0.206 0.25 0.142 1 0.053 0.325 0.684 0.124 0.132 0.292 0.033 -0.027 0.23 0.125 0.201 -0.009 0.171 0.277 0.247 0.27 0.093 0.245

lkdj -0.207 0.039 0.372 -0.142 -0.385 0.026 0.316 0.262 0.053 1 -0.14 -0.021 -0.011 0.27 -0.136 0.175 -0.263 -0.201 -0.124 -0.076 -0.281 0.278 -0.033 0.213 -0.035 0.212 0.114

lnin 0.509 0.313 -0.244 0.448 0.616 0.466 0.017 0.047 0.325 -0.14 1 0.316 0.42 -0.225 0.445 -0.378 0.361 0.405 0.455 0.252 0.178 -0.155 0.256 -0.05 0.386-0.332 0.386

lnwo 0.116 0.11-0.107 0.466 0.353 0.062 0.152 0.159 0.684 -0.021 0.316 1 -0.023 -0.021 0.363 -0.136 0.084 0.243 0.158 0.169 0.306 0.089 0.28 0.114 0.191-0.077 0.168

lpcs 0.564 0.395 0.245 0.318 0.45 0.625 0.11 0.445 0.124 -0.011 0.42 -0.023 1 0.174 0.246 -0.02 0.361 0.458 0.39 0.264 -0.189 0.037 0.454 0.104 0.308 0.046 0.274

lplb -0.022 0.051 0.876 -0.066 -0.111 0.168 0.799 0.556 0.132 0.27 -.225 -0.021 0.174 1-0.276 0.841 -0.056 0.053 -0.21 -0.007 -0.441 0.814 0.275 0.809 -0.198 0.85 -0.132

lpmy 0.611 0.55 -0.395 0.094 0.625 0.315 -0.266 0.163 0.292 -0.136 0.445 0.363 0.246 -0.276 1 -0.357 0.365 0.223 0.729 -0.097 0.099 -0.396 -0.092 -0.312 0.678-0.463 0.511

lpnd -0.136 -0.044 0.828 -0.171 -0.183 0.024 0.74 0.495 0.033 0.175 -.378 -0.136 -0.02 0.841-0.357 1 -0.057 -0.132 -0.286 -0.008 -0.432 0.768 0.073 0.779 -0.269 0.904 -0.209

lrcy 0.301 0.224 -0.087 0.321 0.565 0.558 0.018 0.244 -0.027 -0.263 0.361 0.084 0.361 -0.056 0.365 -0.057 1 0.428 0.299 0.107 -0.113 -0.013 0.252 -0.023 0.241-0.111 0.064

lrkr 0.374 0.05 -0.028 0.592 0.502 0.33 0.028 0.235 0.23 -0.201 0.405 0.243 0.458 0.053 0.223 -0.132 0.428 1 0.244 0.395 0.025 0.143 0.53 -0.02 0.062-0.132 0.037

lzia 0.811 0.778 -0.219 -0.035 0.59 0.376 -0.198 0.149 0.125 -0.124 0.455 0.158 0.39 -0.21 0.729 -0.286 0.299 0.244 1 -0.044 -0.083 -0.372 -0.001 -0.234 0.87-0.385 0.641

2aza 0.115 -0.059 0.001 0.564 0.18 0.314 0.085-0.097 0.201 -0.076 0.252 0.169 0.264 -0.007 -0.097 -0.008 0.107 0.395 -0.044 1 0.124 -0.028 0.192 -0.005 -0.036 0.079 0.039

2cbp -0.114 -0.399 -0.491 0.339 0.037 -0.332 -0.306-0.219-0.009 -0.281 0.178 0.306 -0.189 -0.441 0.099 -0.432 -0.113 0.025 -0.083 0.124 1-0.349 0.102 -0.396 -0.171-0.501 -0.255

2plt -0.186 -0.18 0.786 0.137 -0.171 0.004 0.883 0.638 0.171 0.278 -.155 0.089 0.037 0.814 -0.396 0.768 -0.013 0.143 -0.372 -0.028 -0.349 1 0.41 0.902 -0.455 0.81 -0.23

4azu 0.264 -0.035 0.246 0.446 0.173 0.364 0.427 0.509 0.277 -0.033 0.256 0.28 0.454 0.275 -0.092 0.073 0.252 0.53 -0.001 0.192 0.102 0.41 1 0.317 -0.151 0.143 -0.041

7pcy -0.036 -0.048 0.79 0.092 -0.075 0.053 0.936 0.66 0.247 0.213 -0.05 0.114 0.104 0.809 -0.312 0.779 -0.023 -0.02 -0.234 -0.005 -0.396 0.902 0.317 1 -0.33 0.812 -0.222

8paz 0.736 0.853 -0.212 -0.086 0.499 0.374 -0.291 0.08 0.27 -0.035 0.386 0.191 0.308 -0.198 0.678 -0.269 0.241 0.062 0.87 -0.036 -0.171 -0.455 -0.151 -0.33 1 -0.389 0.735

9pcy" -0.2 -0.113 0.877 -0.026 -0.173 0.001 0.734 0.514 0.093 0.212 -.332 -0.077 0.046 0.85 -0.463 0.904 -0.111-0.132 -0.385 0.079 -0.501 0.81 0.143 0.812 -0.389 1-0.275

aminmr 0.531 0.685 -0.131 -0.057 0.252 0,35 -0.189-0,056 0,245 0.114 0.386 0.168 0.274 -0.132 0.511 -0.209 0.064 0.037 0.641 0.039 -0.255 -0.23 -0.041 -0.222 0.735 -0.275 1

a Reduced form of the plastocyanin from P. Vulgaris.

Fig. 7. (A) Comparison between the cupredoxin pairwise sequence identity percentages [ID(a,b)] and the hydrophobic potential similarity indices (HYD SI). The linear regression is ID(a,b) = 94.93SIHYD + 10.90; n = '2i=i26i; r2 = 0.61; r = 0.78. (B) Comparison between the cupredoxin pairwise sequence identity percentages [ID(a,b)] and the electrostatic potential similarity indices (MEP SI). The linear regression is ID(a,b) = 16.24SImep + 23.58; n = 2,=i26 i; r2 = 0.17; r = 0.41.

Fig. 7. (A) Comparison between the cupredoxin pairwise sequence identity percentages [ID(a,b)] and the hydrophobic potential similarity indices (HYD SI). The linear regression is ID(a,b) = 94.93SIHYD + 10.90; n = '2i=i26i; r2 = 0.61; r = 0.78. (B) Comparison between the cupredoxin pairwise sequence identity percentages [ID(a,b)] and the electrostatic potential similarity indices (MEP SI). The linear regression is ID(a,b) = 16.24SImep + 23.58; n = 2,=i26 i; r2 = 0.17; r = 0.41.

bic patch in the Cu-site region, which is common to all of the cupredoxins (consistent with prior literature). Given the possible importance of hydrophobic features for the electron-transfer process, it might be suggested that all of the cupredoxins should show similarity in electron-transfer properties.

On the other hand, the analysis of MEPs proved to be more interesting and to provide additional information, complementary to that of sequence and structure analyses.

2.4. MIF Similarity Analysis Results

The results showed how the members of the subfamilies can be classified according to their recognition properties, providing clues for experiments to identify redox partners for some of the blue copper proteins.

A conserved hydrophobic region around the Cu site probably plays a functional role in all of the cupredoxins independently from their redox partners. Considering that the Cu is the redox center and is responsible for the transfer of the electrons, it is highly probable that the reason why the northern site is conserved is that it mediates electron transfer.

The other two regions analyzed, the eastern and the MADH/amicyanin site, do not appear to be recognition sites common to all the cupredoxins. The eastern site is probably important only for the plastocyanins, whereas the other is only for the amicyanins.

The low specificity of binding displayed by many of the blue copper proteins is also highlighted by analysis of MEPs. Briefly, the most noteworthy results are as follows:

1. Plastocyanins can be divided, according to their interaction properties, into three subclusters: the eukary-otic plastocyanins that have a highly negative potential becauseof two conserved acidic regions (res. 4245 and 59-61) in the eastern site (Figs. 1 and 5 [top]); the cyanobacterialplastocyanins that, because of their less negative potential (Fig. 5 [top]) and the presence of a unique conserved patch of acidic amino acids (res. 59-61), seem to be more similar to azurins and pseudoazurins; and the fern plastocyanin that shows unique recognition features that are roughly intermediate between those of cyanobacterial and plant plastocyanins.

2. Azurins (from A. xylosoxydans) and pseudoazurins (from A. cycloclastes and A. faecalis), which interact with redox partners belonging to the same family (Cu-containing NIRs) in their reduced form (8,35), show very similar interaction properties in the region around their Cu site. This supports the idea that the binding of these proteins to their partners involves their Cu sites.

3. Amicyanins cluster together with pseudoazurins in MIF similarity space (Fig. 6) rather than with plastocyanins as seen in sequence and structure classifications (Fig. 4). These findings provide new support to the hypothesis (36) that these two proteins (amicyanin and pseudoazurin) can operate as isofunctional proteins under particular environmental conditions.

Finally, it is important to stress that the experimental data available on the binding specificities and the redox partners of the cupredoxins, which are few and often controversial, often do not point to unequivocal functions for these proteins. Therefore, comparative analysis based on similarity indices is a useful aid to unraveling the experimental information, and explaining, on a structural basis, the complex behavior of the cupredoxins.

Automated and reliable techniques, such as the method presented here based on the use of descriptors, are particularly suited for large-scale analysis. This is becoming more and more important in the postgenomic era for structural and functional proteomics, because analyses and comparisons of hundreds of proteins (either structurally determined or modeled) will be required.

0 0

Post a comment